Diploma in Uplift Modeling and Causality-Based Decision Making
Online postgraduate diploma of 800 hours in uplift modelling and causality-based decision-making. 8 modules covering potential outcomes and causal graphs, experimental design, meta-learners, causal trees and forests, Qini evaluation, quasi-experimental methods and budget-constrained treatment policies in line with the GDPR, the EU AI Act and CONSORT 2010, with an assigned technical tutor.
Diploma in Uplift Modeling and Causality-Based Decision Making — studying from South Africa
South Africa has a growing digital economy, with strong demand for data scientists, analysts and simulation specialists in finance, mining, telecoms and government. Major banks, mining houses and telecommunications operators such as MTN and Vodacom rely on advanced analytics for planning and operations. Universities including the University of Cape Town and Wits contribute research in machine learning and computational modelling. Protection of personal data is governed by national legislation, which shapes how analytics work is carried out.
Fees for South Africa: R 24 000 (South African rand), with instalments and employer sponsorship available.
Relevant bodies in South Africa
Typical roles
Questions from South Africa
Is a SEIUM qualification recognised by any South African professional body for data and AI?
South Africa has no single statutory register for data scientists or AI practitioners, and bodies such as the Computer Society of South Africa and the South African Qualifications Authority set their own standards. SEIUM is a private online institute whose qualifications are not state degrees and are not accredited by a national body. Recognition is decided by employers or the relevant body, so ask them directly before enrolling.
Do South African employers in data and AI value private online qualifications?
Demand for data and analytics skills in South Africa spans banking, insurance, telecoms, retail and government, and employers often assess skills through portfolios, technical interviews and experience. A private online qualification may support your CV, but it does not guarantee a job or salary outcome. Show prospective employers the syllabus and your project work to gauge their interest.
How does the workload fit with a job, and what English level do I need?
The programme is fully online and taught in English, so you need confident reading and writing skills. As it is remote, no student visa is needed to study from home. Master's programmes total 1,900 hours, postgraduate diplomas 800 hours and specialist courses 300 hours, with an assigned technical tutor and a final project. Fees can be paid in instalments, and employer sponsorship may be possible.
Take the controls.
Run a real engineering trade-off from this field. Move the sliders, watch the model react in real time and beat the challenge.
What you will be able to do
Frame decision problems in terms of potential outcomes and causal graphs.
Design randomised experiments with adequate power and control groups.
Estimate heterogeneous effects with meta-learners and causal forests.
Evaluate uplift models with Qini curves, AUUC and policy value.
Apply quasi-experimental methods when randomisation is not possible.
Optimise treatment allocation under budget constraints.
Communicate incremental impact rigorously and ethically.
Who it's for
- Data scientists working on causal problems.
- Marketing analytics analysts who need to measure true incremental impact.
- Experimentation leads.
- Machine learning engineers.
- Pricing and promotions consultants.
- Business decision analysts.
- Engineering or science graduates seeking to enter e-commerce.
What you'll work with
- Regulation (EU) 2016/679 (GDPR)
- Regulation (EU) 2024/1689 (EU Artificial Intelligence Act)
- Organic Law 3/2018 (LOPDGDD, Spanish data protection and digital rights law)
- ISO/IEC 22989:2022 (AI concepts and terminology)
- ISO/IEC 23053:2022 (Framework for AI systems using machine learning)
- ISO/IEC 25012:2008 (Data quality model)
- CONSORT 2010 Statement (Reporting of randomised trials)
- CausalML
- EconML
- DoWhy
- grf (R)
- scikit-uplift
8 modules
1.1Neyman-Rubin potential outcomes model1.2ATE, ATT and CATE effects1.3SUTVA, ignorability and positivity assumptions1.4Directed acyclic graphs and d-separation1.5Backdoor criterion and confounder adjustment1.6Simpson's paradox and selection bias1.7Randomised experiments versus observational data1.8Counterfactuals in Pearl's hierarchy1.9Identifiability and the rules of do-calculus1.10Workshop: DAG of a retention campaign with DAGitty
2.1Simple, stratified and block randomisation2.2Sample size and power calculation2.3Holdout groups and contamination between arms2.4Multivariate and factorial A/B designs2.5CUPED and variance reduction2.6Interference and cluster randomisation2.7Sequential tests and early stopping2.8Pre-registration and experiment logging2.9Ethics of experimentation with customers2.10Practical exercise: coupon test with a holdout group
3.1S-learner and regularisation bias3.2T-learner with independent models3.3X-learner for unbalanced treatments3.4R-learner and Robinson orthogonalisation3.5Doubly robust DR-learner3.6Choosing the base model with gradient boosting3.7Propensity estimation and weight trimming3.8CATE-specific cross-validation3.9Comparison of learners on synthetic data3.10Lab: meta-learners with CausalML on the Hillstrom dataset
4.1Uplift trees with KL divergence and chi-squared4.2Honest causal trees by Athey and Imbens4.3Generalised random forests4.4Confidence intervals for the CATE4.5Variable importance in heterogeneity4.6Chernozhukov double machine learning4.7Orthogonal estimators in EconML4.8Jaskowski and Jaroszewicz class transformation4.9Interpreting effects with SHAP values4.10Lab: causal forest with EconML and grf
5.1Uplift curves and Qini curves5.2Qini coefficient and AUUC5.3Incremental gain charts by decile5.4Out-of-sample and temporal stability5.5IPW-weighted metrics with observational data5.6Policy value and off-policy evaluation5.7Calibration of estimated effects5.8Common mistakes when evaluating uplift5.9Comparison with response propensity models5.10Practical exercise: Qini curves for an email campaign
6.1Propensity score matching6.2Inverse probability weighting6.3Difference-in-differences estimator6.4Synthetic control method6.5Instrumental variables and LATE6.6Regression discontinuity designs6.7Sensitivity analysis and E-value6.8Refuting estimates with DoWhy6.9Censoring and missing data in effects6.10Case study: impact of a loyalty programme
7.1Optimal allocation under budget constraints7.2Persuadable, sure thing, lost cause and do-not-disturb segments7.3Multiple treatments with different costs7.4Contextual bandits and Thompson sampling7.5Policy learning with decision trees7.6Fairness and discrimination in allocation7.7Deployment and monitoring of causal drift7.8Incremental return and ROI calculation7.9Communicating results to management7.10Case study: discount allocation with positive ROI
8.1Defining the action and the outcome8.2Data preparation and treatment variable8.3Experimental or observational strategy8.4Training and comparing learners8.5Qini evaluation and policy value8.6Sensitivity analysis of the estimate8.7Deployment and monitoring plan8.8Quantified business recommendation8.9Report and presentation8.10Final project: final review with the tutor
How you'll learn
100% online training on the SEIUM campus, with downloadable material, recorded classes and an assigned technical tutor. There are no fixed timetables: you progress at your own pace within each intake, so the programme is compatible with work.
Each module combines theory, guided reading of the original standard and applied exercises. The working software is CausalML, EconML, DoWhy, grf (R) and scikit-uplift, in free, academic or evaluation versions. Practical work is based on real procedures for retention campaigns with a holdout control group, coupon allocation in e-commerce and impact evaluation with historical CRM data. Assessment is continuous by module plus the final project.
Standards covered: Regulation (EU) 2016/679, Regulation (EU) 2024/1689, Organic Law 3/2018, ISO/IEC 22989:2022, ISO/IEC 23053:2022, ISO/IEC 25012:2008, CONSORT 2010 Statement.
Build it end to end
Your final project follows the real engineering workflow, step by step, reviewed by your tutor.
Where this takes you
Requirements
- Prior technical training or professional experience in the sector. No university degree is required.
- Internet connection and a computer capable of running the programme software.
- English sufficient to read technical standards.
Ready to start?
Send your application today. Our admissions team replies within 72 hours with your study plan and funding options.
Frequently asked questions
The programme totals 800 hours over 6–8 months. You set your own pace within each intake, and your tutor helps you plan the workload around your job.
You work with CausalML, EconML, DoWhy, grf (R) and scikit-uplift, using free, academic or evaluation versions. Your tutor indicates the free alternative wherever one exists.
A SEIUM Institute diploma, awarded on passing the continuous assessment and the final project. It is SEIUM's own qualification, not an official state degree; its recognition depends on the programme completed and the regulations of each country.
Yes. Modules passed are recognised in a higher programme in the same area (postgraduate diploma or master's). Ask about the recognition policy when you enrol.
Prior technical training or professional experience in the sector (no university degree required), a computer able to run the programme software and English sufficient to read technical standards.
Start dates are set by SEIUM's faculty for each intake. Our admissions team will confirm the next available date when you apply.
Related programmes
Diploma in SIL/HIL: Plant Integration and Control
Online postgraduate diploma of 800 hours in SIL/HIL plant and control integration: 8 modules covering in-the-loop testing and the V-model, real-time plant modelling, software-in-the-loop testing, hardware-in-the-loop…
Diploma in Uncertainty Quantification and Propagation
Online postgraduate diploma of 800 hours in uncertainty quantification and propagation: 8 modules covering uncertainty in engineering models, input characterisation, sampling-based propagation, spectral and non-intrusive methods,…
Diploma in Flow and Capacity Optimization with DT
Online postgraduate diploma of 800 hours in flow and capacity optimisation with digital twins: 8 modules covering flows and capacity in operations, digital twins for optimisation…
More in Data, AI, Simulation & Analytics
All Data, AI, Simulation & Analytics programmes → · How to apply · Fees & financing · Scholarships